Forecast Based on Different Data Types: A Before and After Study

Abstract

The estimation of route and mode choice models is of central importance for traffic forecasts. Models which are estimated from revealed preference (RP) data are often subject to statistical problems and might be of limited use. It is therefore standard practice today, to use stated preference (SP) data to supplement or even replace the RP estimates. The main objective of this research project was to predict the changes in traffic demand due to planned infrastructure projects and service changes, and afterwards to verify the results by an before and after analysis. The most important change was the introduction of tilting trains for the Zurich to Lausanne line via Neuchatel in the Swiss Mittelland. For the mode choice forecasts, three different parameter sets were used: Parameters estimated with revealed preference survey data; Parameters estimated with stated preference survey data, which in turn were based on RP data used for the estimation of the previous parameter set; and Demand elasticities based on aggregate data. With a before and after traffic survey (traffic counts and origin-destination surveys), it was possible to compare the actual and the predicted changes of traffic demand, as well as to test the performance of the different parameter sets. The results showed that the forecasts using the SP-parameters performed best, while the RP and elasticity-based forecasts did not perform much worse, they had some consistent problems, which made their application difficult.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 8 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.